© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 80817
    Prev
    Next
    INDEX
    Explanations

    real

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is highly sensitive to occurrences of the word “real.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    \
    -1.27
    borderRadius
    -1.15
          
    -1.13
    </sup>
    -1.11
    انت
    -1.10
     SAME
    -1.08
     Berücksich
    -1.08
    先は
    -1.08
     erneut
    -1.07
    rfloor
    -1.07
    POSITIVE LOGITS
    Réalisation
    1.45
     atacado
    1.40
    Real
    1.37
    real
    1.37
     Verdad
    1.31
     />,
    1.30
     provocó
    1.29
    getReal
    1.24
     واقعی
    1.23
    最後まで
    1.23
    Activations Density 0.069%

    No Known Activations